Learning-based coexistence for LTE operation in unlicensed bands

Abstract

The use of Long Term Evolution (LTE) in the unlicensed 5 GHz band, referred to as LTE-U, is a promising enhancement to increase the capacity of LTE networks and meet the requirements foreseen for future systems. Nevertheless, coexistence among several LTE-U and/or Wi-Fi systems in the same band is a key technical challenge to be resolved. In this context, this paper focuses on the channel selection functionality for LTE-U enabled cells to decide the most appropriate channel to use for downlink traffic offloading in the unlicensed band. A distributed Q-learning mechanism that exploits prior experience is proposed to support this functionality, thus enabling coexistence with other systems in a smart and efficient way. The behavior of the proposed approach is illustrated in an indoor scenario with small cells from two different operators. A fully decentralized approach, where the channel selection decision-making is performed independently by each small cell in the scenario, is considered to initially assess the potentials of the Q-learning solution. Promising results are obtained revealing that the throughput achieved by the proposed approach can be between 96% and 99% of the optimum ideal achievable throughput.